ruihan/qwen38-flash-next-dgx-spark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • 3059ff285a Merge pull request 'Promote validated 128K Spark baseline and consolidate performance evidence' (#1) from perf/compile-prefix-cache into main main ruihan 2026-09-18 01:52:33 +08:00
  • 7f3464fb05 Finalize migration acceptance summary perf/compile-prefix-cache ruihan 2026-09-18 01:51:17 +08:00
  • 73a88a4e8c Record Spark 128K migration acceptance and archive inventory ruihan 2026-09-18 01:50:30 +08:00
  • e290793f2b Set 128K eager baseline and organize deployment documentation ruihan 2026-09-18 01:33:57 +08:00
  • d7e1e745c3 docs(perf): profile CUDA graph coverage and PLE costs on Spark ruihan 2026-09-18 01:22:13 +08:00
  • f1a8964072 Document long-context limits and host OOM at near-262K input ruihan 2026-09-18 00:03:19 +08:00
  • 48a1c9d7c4 Record successful 2048-token CUDA Graph retest and preserve failure samples ruihan 2026-09-17 23:04:40 +08:00
  • 1e2f48d1a4 Enable validated prefix caching and record DGX Spark optimization benchmarks ruihan 2026-09-17 22:09:58 +08:00
  • 5f3030260e Add reproducible DGX Spark deployment for Qwen3.8 Flash Next NVFP4 ruihan 2026-09-17 16:45:50 +08:00